gms | German Medical Science

Objective: The historically poor prognosis of patients with brain metastases (BM) results in their exclusion from many clinical trials. Unfortunately, this trend persists despite improvements in CNS (central nervous system) control and in overall survival of patients with BM.

Method: This is a retrospective single-institutional analysis between 01.01.2008 and 31.12.2010. Overall and post-BM survival were correlated to therapy modalities applying pre- and post-BM. Univariate associations were analyzed using Kaplan-Meier survival analysis, and the effect of independent predictors were assessed using a multivariate Cox proportional hazards regression.

Results: A total of 430 patients with newly diagnosed lung cancer over the given three years period were studied. Male patients made up 66% of the population. The most common histologic type was adenocarcinoma (ADC) (41%), followed by squamous cell carcinoma (SQC) (21%) and small cell carcinoma (SCLC) (12,4%). Of the total number of patients, 120 (28%) developed BM within a mean follow-up time of 8.9 months (3–48 months). BM were most often found in patients with ADC (55%) and SCLC (25%). In 43,6%, BM led to the diagnosis of primary lung cancer, in 27,3%, BM were synchronously diagnosed (within 3 months), and in 29,1% BM developed within 4 years after the primary tumor diagnosis. Overall survival was 18.4 months, BM significantly reduced survival probability (10.6 months versus 23.6 months with or without BM, p=0.00009). Different treatment modalities for BM significantly influenced post-BM survival (p=0.0003): best supportive care was initialized in 9 patients (0.8 months post-BM survival), 71 received WBRT (whole brain radiotherapy) (7.8 months), BM were neurosurgically resected followed by WBRT in 32 (17.8 months), neurosurgical resection was followed by SRS (stereotactic radiosurgery) in 3 (23.15 months), 3 had neurosurgical resection only (2.0 months) and 1 patient each received SRS alone or followed by WBRT.

Conclusions: Our institutional incidence of BM in lung cancer is as high as 28%, with up to 55% in patients with ADC. The outcome in patients with BM is heterogeneous, but in selected patients a tailored therapy with neurosurgical resection followed by WBRT or SRS leads to equally good survival times as in those patients without BM. Clinical trials, especially for NSLC (non-small lung cancer), therefore need to include patients with BM, not only to be more representative of the majority of patients, but to extract those patients with BM, who will benefit from newer therapeutic strategies.